DCEG staff served on a committee of the National Research Council which reviewed the effects of indoor radon on risk of lung cancer and other diseases. The Committee concluded that there were sufficient data to implicate radon as a cause of lung cancer and radon exposure as a significant public health problem. Contrary suggestions from ecologic studies showing an inverse correlation between county radon levels and lung cancer rates were shown to be likely the result of ecologic bias. Data from studies of residential exposures were consistent with risks extrapolated from studies of underground miners, who are generally exposed to much higher radon levels. Other work confirmed an inverse exposure-rate effect for radon, i.e., for equal total exposure, exposures occurring at a lower rate for a longer time imparts a greater risk than exposures occurring at a higher rate for a shorter time. This effect, however, diminished at low total exposure, such as found in houses. A study of lung cancer and indoor radon is underway in an area of China where many homes have high radon concentrations. A case-control study of 638 children under the age of 15 years with acute lymphoblastic leukemia (ALL) who were registered with the Childrens Cancer Group and 620 matched controls, was carried out to evaluate the association of ALL and residential exposure to magnetic fields generated by nearby power lines. Results provided little evidence that living in homes characterized by high measured magnetic-field levels or by configurations of electrical power lines increases the risk of ALL in children. In addition, the study found no association between ALL and residential radon concentration. Using a subset of these data, an analysis found no relationship between most infant vaccinations (oral poliovirus, diphtheria-tetanus-pertussis, measles-mumps-rubella, and Haemophilus influenzae type b (Hib) vaccines) and subsequent risk of childhood ALL. Findings, however, are suggestive that infants receiving conjugate Hib vaccine may be at reduced risk of childhood ALL. Exposure to chlorination products in water was associated with increased risk of rectal cancer, but not colon cancer, in one study and with increased risk of bladder cancer in a second study, but only among smokers. A case-control study of multiple myeloma and family history of cancer among blacks and whites in the US showed that for both races combined, the risk was significantly elevated for subjects who reported that a first- degree relative had multiple myeloma (OR3.7, 95% CI=1.2-12.0). Increased risk was also associated with afamily history of any hematolympho proliferative cancer(OR=1.7, 95%CI=1.0-28). The study indicated a familial risk of multiple myeloma, but explained little of the race-related differences in incidence rates. A case-control study of prostatecancer and dietary factors among blacks and whites in the US showed that greater consumption of fat from animal sources increased risk for prostate cancer among blacks and for advanced prostate cancer among blacks and whites. A case-control study of oral cancer and use of tobacco and alcohol in Puerto Rico showed that the excess risks for oral cancer in Puerto Rico are largely explained by patterns of alcohol and tobacco use. Smoking filtered, as opposed to non-filtered, cigarettes did not alter risk, while cessation of alcohol and tobacco use appeared to reduce risk only gradually. Methodological studies were undertaken to evaluate exposures by farmers from the application of insecticides to animals and methods of data collection, as part of ongoing epidemiologic studies of farmers. Higher exposures were recorded on outer clothing as compared to under clothing, and determinants of high exposures were identified. Poor work practices were associated with higher direct exposure to the active ingredient in insecticides. A case- control study of invasive cervical cancer was conducted in five areas of the United States to examine serum homocysteine levels. Increased risk of cervical cancer was associated with greater serum homocysteine levels. Odds ratios were comparable when cases were compared to HPV-16 positive and HPV-16 negative controls.